Medusa Pricing Groups

What is it?

Medusa Pricing Groups is a plugin for the Medusa headless commerce system that enables one to group products together for the purposes of price calculation.

Quick Start

1. Install the plugin by running npm i -S medusa-products-pricing-groups.
2. Add the following to medusa-config.js to the plugins array:

   {
       resolve: `medusa-products-pricing-groups`,
       options: {
           enableUI: true
       }
   }

3. Run npx medusa migrations run
4. Start the medusa backend

Once installed, you will see a new entry , Pricing Groups in the sidebar. From here you can create pricing groups, add and remove products (click on the newly created group to manage the products).

The plugin will also add an admin widget to the product details view. This widget will show which pricing groups a product is currently associated with.

What problem does it solve?

Out of the box, Medusa has the capability of creating prices for variants of a product in multiple currencies, at multiple different levels, through the use of min_quantity & max_quantity (only possible via API requests currently,
fingers crossed it appears in the dashboard UI soon).

Pricing flexibility is further enabled by price lists, which allow one to set custom prices depending on the audience.

The problem arises when one has a set of products that when purchased together should each contribute towards the volume discounts of the other products in the cart.

Use Case & Example

Criteria

Let's say we have 10 different t-shirts, in varying colours and sizes.

We have a storefront, which retail customers can access and purchase from. We also have two other audiences that will be using the storefront; small quantity resellers, and high volume resellers.

Each of these audiences have their own terms and price lists respectively.

Each variant of t-shirt has volume price breaks.

Each t-shirt variant can have different prices.

For the sake of the example we will assume the following prices apply:

For retail customers:

• 1-3 t-shirts: 20
• 3+ t-shirts: 18

For the small quantity & high volume reseller audiences:

• 1-5 t-shirts: 20
• 5-10 t-shirts: 15
• 10+ shirts: 12

All t-shirts cost the same in the example, though in practice the prices can differ, and be set as normal.

Requirements

What we are looking to achieve is the following criteria:

• Each product variant can have its prices set independently in all supported currencies
• Products can easily be grouped and ungrouped
• Products can have different volume breakpoints (i.e. min_quantity/max_quantity thresholds can differ between grouped products)
• Any applicable volume discounts should be automatically applied
• If multiple discounts are available for a set of products in the cart the best discount should be chosen.

The Example

We have our example case & criteria set out above.

The following are the expected cart totals:

• A retail customer adds 1 x red & 1 x blue to their cart.
  • Pricing tier: 1-3 t-shirts
  • Cart total: 40 (20 + 20)
• A retail customer adds 1 x red, 1 x green & 2 x purple to their cart.
  • Pricing tier: 3+ t-shirts
  • Cart total: 72 (18 + 18 + 36)
• A small reseller adds 1 x red, 1 x green & 2 x purple to their cart.
  • Pricing tier: 1-5 t-shirts
  • Cart total: 80 (20 + 20 + 40)
• A small reseller adds 4 x blue, 4 x green, 3 x purple to their cart.
  • Pricing tier: 10+ t-shirts
  • Cart total: 132 (48 + 48 + 36)

The problem can be partially solved with promotional codes, though this becomes a complexity nightmare, in terms of tracking active codes, calculating fixed/percentage discounts, monitoring edge cases, and so on. It quickly becomes impossible to achieve pricing structures due to the rigidity of fixed/percentage discounts, whilst also being a poor user experience in terms of the user having to choose & manually apply one of potentially many codes.

How this plugin solves it

The concept of a pricing group is introduced. A pricing group is a very simple structure, comprised of an identifier and a name, along with a set of products tagged with the identifier.

The plugin implements a pricing group service that retrieves these linkages between the pricing group and products.

The plugin implements a pricing strategy that calculates and returns the best price for the variant in question.

In simple terms we do the following:

1. Retrieve all variants in the cart.
2. Iterate all variants, and retrieve associated groups for each variant.
3. Create a map of Variant ID => [Group IDs]
4. Create a map of Group ID => Calculated Quantity
5. Iterate variants in cart, increment Calculated Quantity for each Group ID associated with current Variant ID by Variant Quantity
6. Use Calculated Quantity to retrieve the Best Price available.
7. Calculate line item total as Variant Quantity x Best Price.

N.B. If the cart is undefined for the request, we fall back to retrieving the price for the variant as if it were a standalone product rather than a grouped one.

Development Notes

When developing the package, attempting to run this plugin on a medusa backend will cause issues if you install @medusajs/medusa into the node_modules folder of the plugin directory, due to dependency conflicts with the medusa backend.

To workaround this:

1. cd into node_modules/@medusajs/medusa from the root of the medusa backend directory, and run yarn link.
2. cd into the plugin directory, then run yarn link @medusajs/medusa. The plugin will now build using the @medusajs/medusa dependency from the backend.
3. Run yarn link in the root of the plugin directory.
4. cd into the backend directory again, and run yarn link medusa-products-pricing-groups.
5. You can now run yarn dev to start the backend, make some changes to the plugin, and run yarn build in the plugin directory to see the changes live reloaded on the backend dashboard UI.

Afterthought / Disclaimer

I am fully aware the code quality isn't perfect as things currently stand - this was somewhat of a proof-of-concept project to familiarise myself with Medusa and see if it would suit our needs.

There may be bugs, I haven't fully tested this plugin in production.

Also, the data table elements were lifted from the core admin dashboard and modified, as they more or less suited my needs and I wasn't about to write them from scratch.